Historical Precedent Depth Score

Overview

This pillar assesses the strength and relevance of historical precedents for a given market. It determines if there is enough quality data from the past to form a reliable base rate for prediction.

What It Does

The pillar systematically searches for and analyzes past events that are analogous to the current prediction market. It quantifies the number of similar occurrences (N-count), evaluates the quality of the data, and flags potential issues like survivorship bias. The output is a simple score indicating how much weight you can confidently place on historical patterns.

Why It Matters

Many prediction errors come from relying on a small or irrelevant set of historical examples. This pillar provides a disciplined, quantitative check on the data's foundation, preventing overconfidence and highlighting markets where history is a poor guide.

How It Works

First, the pillar defines the core attributes of the market question. It then queries historical databases to find matching events within a relevant timeframe. Each match is scored for similarity, and the overall dataset is checked for completeness and bias, resulting in a final Depth Score.

Methodology

The core calculation is the Depth Score, derived from: (log(N) * Average Relevance Score) - Bias Penalty. N is the count of similar historical events. The Average Relevance Score (0-1) is based on a weighted average of contextual similarities. A Bias Penalty is applied if survivorship bias or significant data gaps are detected.

Edge & Advantage

It provides a crucial defensive edge by preventing you from making confident positions on markets that lack a statistically significant historical foundation.

Key Indicators

• N-Count of Similar Events

  high

  The raw number of comparable past instances found in historical data.

• Data Quality Index

  high

  A composite score (0-100) rating the reliability, completeness, and verifiability of the historical data.

• Survivorship Bias Flag

  high

  A binary flag that activates if the historical dataset overwhelmingly represents successes while ignoring failures.

Data Sources

• Academic Archives (JSTOR, Google Scholar)

  Provides access to peer-reviewed studies and historical event data across various domains.

• Government Statistical Agencies

  Sources like the Bureau of Labor Statistics or Census Bureau offer robust demographic and economic base rates.

• Financial Data Providers

  Services like Refinitiv or Bloomberg Terminal archives provide deep historical market performance data.

Example Questions This Pillar Answers

• → Will a third-party candidate win more than 5% of the popular vote in the next US presidential election?
• → Will the S&P 500 have a negative return the year after a US presidential election?
• → Will a new social media app reach 100 million users within its first year?

Tags